google_search.go

package main

import (
	"fmt"
	"log"
	"net/http"
	"net/url"
	"strings"

	"github.com/PuerkitoBio/goquery"
)

type Result struct {
	Title string
	Url   string
	Rank  int
	Query string
}

func GoogleSearch(query string) (result []Result, totResult int, filteredResult int) {
	client := &http.Client{}

	req, err := http.NewRequest("GET", "https://www.google.it/search?q="+url.QueryEscape(query), nil)
	if err != nil {
		log.Fatalln(err)
	}

	req.Header.Set("User-Agent", "Mozilla/5.0 (Macintosh; Intel Mac OS X 10_10_4) AppleWebKit/537.36 (KHTML, like Ge    cko) Chrome/44.0.2403.130 Safari/537.36")

	resp, err := client.Do(req)
	if err != nil {
		log.Fatalln(err)
	}

	defer resp.Body.Close()

	doc, err := goquery.NewDocumentFromReader(resp.Body)
	if err != nil {
		log.Fatal(err)
	}

	skip := []string{
		"tripadvisor.it",
		"elenchi.com",
		"youtube.com",
		"reteimprese.it",
		"icitta.it",
		"impresaitalia.info",
		"cercoimprese.com",
		"elenchitelefonici.it",
		"aziendit.com",
		"paginegialle.it",
		"misterimprese.it",
		"facebook.com",
		"paginebianche.it",
		"virgilio.it",
		"telefonare.it",
		"aziendix.com",
		"paginemail.it",
		"trovanumeri.com",
		"infobel.com",
		"plus.google.com",
	}

	doc.Find("h3.r a").Each(func(i int, s *goquery.Selection) {
		totResult += 1
		link, _ := s.Attr("href")
		title := s.Text()
		//fmt.Printf("Review %d: %s - %s\n", i, title, link)
		found := false
		for _, s := range skip {
			if strings.Contains(link, s) {
				found = true
				filteredResult += 1
			}
		}

		if !found {
			result = append(result, Result{title, link, i + 1, query})
		}
	})
	return
}

func main() {
	res, tot, fil := GoogleSearch("query")
	fmt.Println(res)
	fmt.Println(tot)
	fmt.Println(fil)
}